cCorrGAN: Conditional Correlation GAN for Learning Empirical Conditional Distributions in the Elliptope

We propose a methodology to approximate conditional distributions in the elliptope of correlation matrices based on conditional generative adversarial networks. We illustrate the methodology with an application from quantitative finance: Monte Carlo simulations of correlated returns to compare risk-based portfolio construction methods. Finally, we discuss about current limitations and advocate for further exploration of the elliptope geometry to improve results.

[1]  N. Higham.  Computing the nearest correlation matrix—a problem from finance , 2002 .

[2]  Magnus Wiese,et al.  Quant GANs: deep generation of financial time series , 2019, Quantitative Finance.

[3]  Jie Chen,et al.  Time Series Simulation by Conditional Generative Adversarial Net , 2019, International Journal of Neural Networks and Advanced Applications.

[4]  Gautier Marti,et al.  CORRGAN: Sampling Realistic Financial Correlation Matrices Using Generative Adversarial Networks , 2019, ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[5]  Frank Nielsen,et al.  Clustering in Hilbert’s Projective Geometry: The Case Studies of the Probability Simplex and the Elliptope of Correlation Matrices , 2018, Geometric Structures of Information.

[6]  Fisher-Rao distance on the covariance cone , 2020, 2010.15861.

[7]  Philip I. Davies,et al.  Numerically Stable Generation of Correlation Matrices and Their Factors , 2000 .

[8]  Takashi Masuko,et al.  Covariance clustering on Riemannian manifolds for acoustic model compression , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[9]  Jochen Papenbrock,et al.  Matrix Evolutions: Synthetic Correlations and Explainable Machine Learning for Constructing Robust Investment Portfolios , 2020, The Journal of Financial Data Science.

[10]  Dorota Kurowicka,et al.  Generating random correlation matrices based on vines and extended onion method , 2009, J. Multivar. Anal..

[11]  Matthieu Cord,et al.  Riemannian batch normalization for SPD neural networks , 2019, NeurIPS.

[12]  Luc Van Gool,et al.  A Riemannian Network for SPD Matrix Learning , 2016, AAAI.

[13]  Jan-Frederik Mai,et al.  Simulating realistic correlation matrices for financial applications: correlation matrices with the Perron–Frobenius property , 2018, Journal of Statistical Computation and Simulation.

[14]  M. L. Prado.  Building Diversified Portfolios that Outperform Out-of-Sample , 2016 .

[15]  Masashi Sugiyama,et al.  Geometry-aware principal component analysis for symmetric positive definite matrices , 2017, Machine Learning.

[16]  Samuel A. Assefa.  Generating synthetic data in finance: opportunities, challenges and pitfalls , 2020, ICAIF.

[17]  Frank Nielsen,et al.  Matrix Information Geometry , 2012 .

[18]  Shane G. Henderson,et al.  Behavior of the NORTA method for correlated random vector generation as the dimension increases , 2003, TOMC.

[19]  Maher Moakher,et al.  A Differential Geometric Approach to the Geometric Mean of Symmetric Positive-Definite Matrices , 2005, SIAM J. Matrix Anal. Appl..

[20]  Kumiko Tanaka-Ishii,et al.  Modeling financial time-series with generative adversarial networks , 2019, Physica A: Statistical Mechanics and its Applications.

[21]  Philip C. Treleaven,et al.  Generative adversarial networks for financial trading strategies fine-tuning and combination , 2019, Quantitative Finance.

[22]  Building Diversified Portfolios that Outperform Out of Sample , 2016 .

[23]  Christian Jutten,et al.  Classification of covariance matrices using a Riemannian-based kernel for BCI applications , 2013, Neurocomputing.

[24]  Peter T. Kim,et al.  Geodesic Clustering for Covariance Matrices , 2015 .

[25]  Yu Xue,et al.  Generative adversarial network based telecom fraud detection at the receiving bank , 2018, Neural Networks.